9c70c574b4
Summary: - The current implementation simplifies the case where the source of `copyto` is `implicit-def`ed. However, it only works when that `implicit-def` is single-used since it detects that from `implicit-def` and cannot determine which destination vreg should be used if there are multiple uses. - This patch changes that detection when `copyto` is being emitted. If that `copyto`'s source is defined from `implicit-def`, it simplifies it. Hence, it works even that `implicit-def` is multi-used. - Except it simplifies the internal IR, it won't improve the quality of code generation. However, it helps to detect 'implicit-def` in a straight-forward manner in some passes, such as `si-i1-copies`. A test case is added. Reviewers: sunfish, nhaehnle Subscribers: jvesely, hiraditya, asbirlea, llvm-commits, yaxunl Tags: #llvm Differential Revision: https://reviews.llvm.org/D62342 llvm-svn: 361777
23 lines
654 B
LLVM
23 lines
654 B
LLVM
; RUN: llc -march=amdgcn -stop-after=amdgpu-isel -verify-machineinstrs -o - %s | FileCheck %s
|
|
|
|
; CHECK-LABEL: vcopy_i1_undef
|
|
; CHECK: IMPLICIT_DEF
|
|
; CHECK-NOT: COPY
|
|
; CHECK: IMPLICIT_DEF
|
|
; CHECK-NOT: COPY
|
|
; CHECK: .false:
|
|
define <2 x float> @vcopy_i1_undef(<2 x float> addrspace(1)* %p) {
|
|
entry:
|
|
br i1 undef, label %exit, label %false
|
|
|
|
false:
|
|
%x = load <2 x float>, <2 x float> addrspace(1)* %p
|
|
%cmp = fcmp one <2 x float> %x, zeroinitializer
|
|
br label %exit
|
|
|
|
exit:
|
|
%c = phi <2 x i1> [ undef, %entry ], [ %cmp, %false ]
|
|
%ret = select <2 x i1> %c, <2 x float> <float 2.0, float 2.0>, <2 x float> <float 4.0, float 4.0>
|
|
ret <2 x float> %ret
|
|
}
|